Various data services select or recommend content for display to users. For example, in the self-help setting, a variety of data services provide tips, recommendations, and focused content to assist a subject human user with goal-based outcomes such as exercise goals, weight loss, smoking cessation, medical therapy, and the like. Some of these data services provide recommended content to a user in response to user-indicated preferences, user-indicated activity history, or manual user requests for content. Other data services rely on an expert human user to determine which content is most appropriate for delivery to the subject human user to achieve a goal-based outcome.
To the extent that existing data services provide automated recommendations or selections of content, the timing, delivery, and substance of content from these data services is determined by complex predetermined rules and attributes, or other selections influenced by manual human intervention. For example, recommendations may be hard-coded in a content delivery system to deliver suggestive content at scheduled intervals, or in response to the user's manual indications that a certain accomplishment has or has not been reached. Existing systems and techniques do not provide real-time recommendations and content selections without extensive programming or oversight. Further, the workflows involved with existing content delivery systems call for extensive human selection and specification of content, and are not fully automated, responsive, or adaptive to user needs.